INDEX

Explanations

process

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 process

-1.27

 plan

-1.12

process

-1.03

 Process

-0.82

Process

-0.77

 PROCESS

-0.72

plan

-0.69

 proceso

-0.67

 processes

-0.65

PROCESS

-0.64

POSITIVE LOGITS

homonymie

0.67

AndEndTag

0.67

 jspb

0.66

 electing

0.63

 overruled

0.62

 thanking

0.60

 elsewhere

0.60

 unanimously

0.60

TypedDataSet

0.60

الدراسه

0.58

Activations Density 0.035%